www.nature.com/bjc

ARTICLE Molecular Diagnostics papilloma virus (HPV) integration signature in Cervical Cancer: identification of MACROD2 as HPV hot spot integration site

Maud Kamal1,2, Sonia Lameiras3, Marc Deloger4, Adeline Morel5, Sophie Vacher5, Charlotte Lecerf1,2, Célia Dupain1,2, Emmanuelle Jeannot 5,6, Elodie Girard4, Sylvain Baulande3, Coraline Dubot1,2, Gemma Kenter7, Ekaterina S. Jordanova 7,8, Els M. J. J. Berns9, Guillaume Bataillon6, Marina Popovic10, Roman Rouzier11,12, Wulfran Cacheux13, Christophe Le Tourneau 1,2,4,12, Alain Nicolas14, Nicolas Servant4, Suzy M. Scholl1,2, Ivan Bièche5,15 and RAIDs Consortium

BACKGROUND: Cervical cancer (CC) remains a leading cause of gynaecological cancer-related mortality with infection by human papilloma virus (HPV) being the most important risk factor. We analysed the association between different viral integration signatures, clinical parameters and outcome in pre-treated CCs. METHODS: Different integration signatures were identified using HPV double capture followed by next-generation sequencing (NGS) in 272 CC patients from the BioRAIDs study [NCT02428842]. Correlations between HPV integration signatures and clinical, biological and molecular features were assessed. RESULTS: Episomal HPV was much less frequent in CC as compared to anal carcinoma (p < 0.0001). We identified >300 different HPV-chromosomal junctions (inter- or intra-genic). The most frequent integration site in CC was in MACROD2 gene followed by MIPOL1/TTC6 and TP63. HPV integration signatures were not associated with histological subtype, FIGO staging, treatment or PFS. HPVs were more frequently episomal in PIK3CA mutated tumours (p = 0.023). Viral integration type was dependent on HPV genotype (p < 0.0001); HPV18 and HPV45 being always integrated. High HPV copy number was associated with longer PFS (p = 0.011). CONCLUSIONS: This is to our knowledge the first study assessing the prognostic value of HPV integration in a prospectively annotated CC cohort, which detects a hotspot of HPV integration at MACROD2; involved in impaired PARP1 activity and instability.

British Journal of Cancer (2021) 124:777–785; https://doi.org/10.1038/s41416-020-01153-4

BACKGROUND genome. Both patterns may be present jointly (episomal/ Cervical cancer (CC) remains a leading cause of gynaecological integrated).5 It is thought that the longer half-life of integrated cancer-related mortality worldwide and constitutes the second viral transcripts compared to half-life of episomal transcripts most common malignancy in women.1 Although patients with CC favours cellular immortalisation and transformation into cancer exhibit differences in clinical behaviour, infection by high-risk cells while also providing a selective growth advantage.6 Most human papilloma virus (HPV) remains an important initiating often, the integration of HPV DNA leads to a breakpoint in the E2 event in CC tumorigenesis,2 and one of the most important risk gene, resulting in de-repression of the E6 and E7 viral oncogenes. factors for developing CC.3 Most HPV infections are cleared When the virus remains episomal, expression of E6 and E7 spontaneously by the , yet in some cases, it may result from leaky expression or epigenetics persists leading to cancer.4 Following infection, the virus can dysregulation. E6 and E7 proteins impact the function of p53 remain in its episomal form, or become integrated into the host and pRb proteins, allowing squamous cell tumorigenesis.6

1Department of Drug Development and Innovation, Institut Curie, PSL Research University, 75005 Paris & 92210, Saint-Cloud, France; 2Department of Drug Development and Innovation, Institut Curie, PSL Research University, 92210 Saint-Cloud, France; 3Institut Curie, Genomics of Excellence (ICGex) Platform, PSL Research University, 75005 Paris, France; 4Bioinformatics and Computational Systems Biology of Cancer, PSL Research University, Mines Paris Tech, INSERM U900, 75005 Paris, France; 5Department of Genetics, Institut Curie, PSL Research University, 75005 Paris, France; 6Department of Pathology, Institut Curie, PSL Research University, 75005 Paris, France; 7Center for Gynaecologic Oncology Amsterdam, Amsterdam UMC and The Netherlands Cancer Institute - Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands; 8Department of Pathology, Leiden University Medical Center, Leiden, The Netherlands; 9Department of Medical Oncology, Erasmus MC, 3000 CA Rotterdam, The Netherlands; 10Oncology Institute of Vojvodina, Put doktora Goldmana, 421204 Sremska Kamenica, Serbia; 11Department of Surgery, Institut Curie, PSL Research University, 92210 Saint-Cloud, France; 12Paris-Saclay University, Paris, France; 13Hopital Privé Pays de Savoie, Service d’oncologie médicale, 19 avenue Pierre Mendès France, 74100 Annemasse, France; 14Institut Curie, PSL Research University, CNRS UMR3244, 75248 Paris, France and 15Faculty of Pharmaceutical and Biological Sciences, INSERM U1016, Paris Descartes University, 75005 Paris, France Correspondence: Maud Kamal ([email protected]) Members of the RAIDs Consortium are listed end of the paper. These authors contributed equally: Suzy M. Scholl, Ivan Bièche. Received: 18 March 2020 Revised: 2 October 2020 Accepted: 15 October 2020 Published online: 16 November 2020

© The Author(s) 2020 Published by Springer Nature on behalf of Cancer Research UK Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 778 Several mechanisms of integration have been reported in the Table 1. Clinical and biological characteristics of 272 patients with literature; the “looping” model of HPV integration following DNA HPV-positive cervical cancer, in relation to progression-free survival. replication and recombination (resulting in DNA concatemers)7 is the most widely accepted but not experimentally reconstituted. Patients (%) Events (%)a PFS (p HPV DNA integration into the triggers various value)b genetic alterations, such as oncogenes amplification, tumour suppressor gene inactivation, inter- or intra- chromosomal Total 272 (100.0) 84 (30.9) 6,8 rearrangements as well as genetic instability. localised Age near the integration sites of viral genomes can experience ≤50 140 (51.5) 41 (29.3) 0.40 (NS) changes in RNA and expression levels, leading to over- >50 132 (48.5) 43 (32.6) or under-expression. In 2015, whole-genome sequencing and high-throughput viral integration methods identified as many as Histologic subtype 3667 HPV integration breakpoints in cervical neoplastic lesions. Squamous cell carcinoma 230 (84.6) 71 (30.9) 0.56 (NS) Frequent integration sites have been reported in genes relevant to Adenocarcinoma 27 (9.9) 8 (29.6) 9 the neoplastic process, such as the MYC oncogene. Loss of Adenosquamous 10 (3.7) 4 (40.0) function (LOF) in the RAD51B tumour suppressor gene following carcinoma HPV DNA insertion was reported to affect the DNA repair pathway 10 Mixed form or 5 (1.8) 1 (20.0) and genomic instability in CC. undifferentiated HPV DNA integration occurs as a single copy or in multiple HPV status repeats (in tandem or dispersed).11 In 2016, Holmes et al. developed a Capture HPV method to identify five different HPV Genotype 16 155 (57.0) 50 (32.3) 0.13 (NS) signatures in 72 CC. The first two signatures contain two hybrid Genotype 18 36 (13.2) 14 (38.9) chromosomal–HPV junctions which are co-linear (2 Junctions Genotype 45 27 (9.9) 10 (37.0) “ ” “ ” Colinear 2J-COL ) or non-linear (2 Junctions Non-Linear 2J-NL ) Genotype 31 9 (3.3) 0 (0) depending on their relative orientations. It reflects two modes of Genotype 33 11 (4.0) 0 (0) viral integration, associated with chromosomal deletion or c amplification events, respectively. The third and fourth signatures Other genotypes 34 (12.5) 10 (29.4)

1234567890();,: exhibit several hybrid junctions either clustered in one chromo- FIGO stage somal region (Multiple Junctions Clustered “MJ-CL”) or scattered at I/II 205 (75.4) 50 (24.4) <0.0001 “ ” fi distinct loci (Multiple Junctions Scattered MJ-SC ) while the fth III/IV 67 (24.6) 34 (50.7) signature consists of episomal forms of HPV (EPI).12 Nodal involvement On the assumption that HPV integration types/signatures/ pattern might predict clinical outcomes, we analysed the Yes 167 (61.4) 62 (37.1) 0.0028 association between the different viral integration signatures, No 105 (38.6) 22 (21.0) clinical and pathological parameters and outcome in the large Pelvic lymph nodes cohort of 272 HPV-positive CC patients enrolled in the prospective Yes 165 (60.7) 62 (37.6) 0.0015 BioRAIDs study [NCT02428842]. No 107 (39.3) 22 (20.6) Para-aortic lymph nodes METHODS Yes 43 (15.8) 22 (51.2) 0.0001 Patients and samples No 229 (84.2) 62 (27.1) Patients included in this study were enrolled in the EU-funded Initial therapy RAIDs Network (Rational Molecular Assessment and Innovative Surgery 54 (19.9) 10 (18.5) 0.0008 Drug Selection, www.raids-fp7.eu) prospective CC BioRAIDs study [NCT02428842]. The clinical protocol together with tumour Radiotherapy 176 (64.7) 52 (29.5) sampling procedures, quality control of samples and treatment Neoadjuvant 42 (15.4) 22 (52.4) in 18 European centres (seven European countries) as well as chemotherapy – study results have been previously published.13 15 PIK3CA mutational statusd WT 182 (67.7) 60 (33.0) 0.23 (NS) HPV typing Mutated 87 (32.3) 23 (26.4) All samples included in this study were analysed for HPV type, using the SPF10 primer set and INNO-LiPA HPV genotyping extra Significant results are displayed in bold. line probe assay (Fujirebio Europe, Gent, Belgium) according to NS not significant. ’ fi μ aUntil 24 months. the manufacturers protocol. For DNA isolation, one to ve 10 m b tissue sections were cut depending on the size of the tumour Log-rank test. cOther HPV genotypes: 39, 42, 52, 56, 58, 59, 68, 70, 73, 82. biopsy. DNA was isolated using the automated Tissue Preparation dInformation available for 269 patients. System (Siemens Healthcare Diagnostics, NY, USA).

PIK3CA detection A mutational analysis of the PIK3CA gene had been previously COSMIC database were considered. Among the 87 PIK3CA carried out on all tumour samples.15 In summary, paired-end , three patients had an H1047R mutation ( 20) whole-exome sequencing was performed on a HiSeq2500 plat- and 84 patients had a E452K/E545K mutation (exon 9). form, with an Agilent SureSelectXT Human. The sequencing was performed to reach an average depth of coverage of at least 80× DNA library preparation per sample. Dedicated filtering strategies were applied to somatic The DNA libraries were prepared using 500 ng of genomic DNA variants depending on their functional impact per gene category: (extracted from frozen tissue), starting with ultra-sonication oncogene or tumour suppressor gene or uncharacterised. For (Covaris) to produce double-strand DNA fragments of approxi- oncogenes as PIK3CA, hotspot missense mutations known in the mately 280 bp. End-Repair and A-tailing were applied to facilitate Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 779 ligation of the adapters, containing unique barcodes for each genotypes. The DNA sequences were then captured by streptavidin sample, specific to the Illumina technology for amplification and beads and amplified by PCR. We performed a double capture (i.e. sequencing. KAPA Hyper Prep kit was used, according to the two rounds of hybridisation and capture) to improve the efficiency manufacturer’s instructions. and specificity. Post-capture libraries were sequenced using Illumina MiSeq system (Illumina, San Diego, CA, USA), in paired-end 150, with HPV double capture method 24 samples multiplexed on a V2 micro flow-cell. The double capture method was carried out using the SeqCap EZ The HPV copy number shows the abundance of the target Rapid Library Small Target Capture method, developed by Roche, relative to the endogenous control (KLK3) in order to normalise which is adapted to capture small DNA targets. The DNA libraries the starting amount and quality of genomic DNA. Similar results were multiplexed (by 12) and hybridised for 16 h with the were obtained with other endogenous diploid controls (GAPDH, biotinylated HPV oligonucleotide probes, recognising all HPV RAB7A).

Table 2. Relationship between mechanisms of integration of HPV and clinical, biological and pathological characteristics of the 272 patients with HPV-positive cervical cancer.

Number of patients (%) HPV insertiona Patients (%) EPI 2J MJ p valueb

Total 272 (100.0) 33 (12.1) 117 (43.0) 122 (44.9) Age ≤50 140 (51.5) 11 (33.3) 60 (51.3) 69 (56.6) 0.061 (NS) >50 132 (48.5) 22 (66.7) 57 (48.7) 53 (43.4) Histologic subtype Squamous cell carcinoma 230 (84.6) 27 (81.8) 92 (78.6) 111 (91.1) 0.14 (NS) Adenocarcinoma 27 (9.9) 3 (9.1) 18 (15.4) 6 (4.9) Adenosquamous carcinoma 10 (3.7) 2 (6.1) 4 (3.4) 4 (3.3) Mixed form or undifferenciated 5 (1.8) 1 (3.0) 3 (2.6) 1 (0.8) HPV status Genotype 16 155 (57.0) 25 (75.8) 44 (37.6) 86 (70.5) <0.0001 Genotype 18 36 (13.2) 0 (0.0) 24 (20.5) 12 (9.8) Genotype 45 27 (9.9) 0 (0.0) 24 (20.5) 3 (2.5) Genotype 31 9 (3.3) 2 (6.1) 4 (3.4) 3 (2.5) Genotype 33 11 (4.0) 3 (9.1) 4 (3.4) 4 (3.3) Other genotypesc 34 (12.5) 3 (9.1) 17 (14.5) 14 (11.5) Stage FIGO I/II 205 (75.4) 25 (75.8) 86 (73.5) 94 (77.0) 0.82 (NS) III/IV 67 (24.6) 8 (24.2) 31 (26.5) 28 (23.0) Lymph node Yes 167 (61.4) 19 (57.6) 81 (69.2) 67 (54.9) 0.067 (NS) No 105 (38.6) 14 (42.4) 36 (30.8) 55 (45.1) Pelvis lymph node Yes 165 (60.7) 19 (57.6) 80 (68.4) 66 (54.1) 0.072 (NS) No 107 (39.3) 14 (42.4) 37 (31.6) 56 (45.9) Para-aortic lymph node Yes 43 (15.8) 3 (9.1) 20 (17.1) 20 (16.4) 0.52 (NS) No 229 (84.2) 30 (90.9) 97 (82.9) 102 (83.6) Initial therapy Surgery 54 (19.9) 7 (21.2) 21 (17.9) 26 (21.3) 0.86 (NS) Radiotherapy 176 (64.7) 22 (66.7) 75 (64.1) 79 (64.8) Neoadjuvant chemotherapy 42 (15.4) 4 (12.1) 21 (17.9) 17 (13.9) PIK3CA mutational statusd WT 182 (67.7) 15 (46.9) 84 (72.4) 83 (68.6) 0.023 Mutated 87 (32.3) 17 (53.1) 32 (27.6) 38 (31.4) Significant results are displayed in bold. NS not significant. aHPV insertion: EPI episomal, 2J 2 junctions, MJ multiple junctions. bChi-square test; p values for comparison of the EPI group vs. the 2J group vs. the MJ group for each parameter. cOther HPV genotypes: 39, 42, 52, 56, 58, 59, 68, 70, 73, and 82. dInformation available for 269 patients. Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 780 ab

EPI 12% MJ-SC 27% MJ-SC 2J-COL EPI 35% 11% 45%

MJ-CL 2J-NL 10% 19% MJ-CL 2J 2J- 10% 9% 2J COL 13% 7% 2J-NL 2% Cervical cancer Anal cancer17 (n=272) (n=93) Fig. 1 Distribution of the HPV integration signatures according to the location of HPV-positive squamous cell carcinomas. a Cervical cancer; b anal cancer; p < 0.0001. 2J-COL 2 Junctions Colinear, 2J-NL 2 Junctions Non-Linear, MJ-CL Multiple Junctions Clustered, MJ-SC Multiple Junctions Scattered, EPI episomal and 2J 2 Junctions.

36.3 36.2 36.1 25 35 24 MACROD2 (n=7) 34.2 33 23 32 22 MIPOL1/TTC6 (n=5) 21 26 31 16 25 TP63 (n=5) 24 14 16 22 15.3 Other sites (n=291) 13 22 15.3 25 15.1 21 12 15.1 24 21 23 22 11.2 14 14 13 13 13 22 21 11.2 14 12 22.3 12 12 21.3 22.2 12 12 15 23 24 13 15 22.1 13 11.2 14 14.1 13 21.2 14 22 23 15 12 12 12 13 13 21 14.3 12 21 21 21 13 12 12 14 11.4 21 21 12 11.2 12 11.2 13 11.2 13 22 22 22 12 11.2 13.1 14 13 11.2 12 23 23 24 14 11.2 15 15 11.2 11.2 24 13.3 12 12 24 21 13 25 26 16 21 13 12 21 22 13 21 31 22 27 21 21.2 31 22 21 13 21 32.1 23 23 28 21.3 22 32.3 24 22 14 22 31.1 31 22 32 33 25 31 22 23 21 23 31.3 23 34 26.1 31 22 41 32 32 24 32 23 24 25 35 26.3 33 33 32 33 33 25 23 26 42 36 27 34 25 35 24.1 43 28 34 26 34 27 37 35 36 26 24 44 29 35 27 24.3 25 28 12345678 91011X

13 12 13 13 11.2 12 12 13 11.2 11.2 12 12 11.2 13.3 12 11.2 13 13 12 13.1 13 12 12 11.3 14 13 14 12 13.3 1 15 11.2 11.2 15 11.2 13.2 13 21 11.2 13 11.3 21 21 11.2 13 12 13 11.2 21 22 12 12 12 12 11.2 11.2 23 22 12.1 21 12 11.2 11.2 22 22 12 24 23 13 11.2 11.2 11.2 23 31 13 22 21 13.1 11.2 24 21 12 24.1 22 23 13.2 12 32 31 25 24 24.2 23 22 13.3 13.1 12 33 26 22 13 24.3 34 32 24 25 23 13.4 13.3 12 13 14 15 16 17 18 19 20 21 22 Y Fig. 2 Distribution of HPV insertion sites in the genome of patients with HPV-positive CC. Each dot represents an HPV integration site.

Bioinformatics analyses already described in Holmes et al. Briefly nf-VIF performs (i) quality In order to analyse our HPV capture data, we set up a new controls and cleaning of raw sequencing Illumina data, (ii) HPV pipeline called nf-VIF available at https://github. genotyping, and (iii) the detection of the HPV insertion sites within com/bioinfo-pf-curie/nf-vif/, which implements the methods we the human genome. Nf-VIF is implemented through the Nextflow Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 781 workflow management system, ensuring a high portability, Table 3. Clinical and biological characteristics of 272 HPV-positive reproducibility, and scalability (see Supplementary materials for cervical cancer, in relation to HPV copy number. details). Number of patients (%) Statistical analysis a The correlations between HPV integration signatures and clinical, HPV copy number Patients (%) Low HPV High HPV copy copy biological and molecular features were analysed using chi-square number tests, chi-square tests with Yates’ correction or Fisher’s exact tests, (<4) ≥ as appropriate. Progression-free survival (PFS) was defined as the number ( 4) p valueb time interval from the date of CC diagnosis to progression. Survival data were censored on the date of last follow-up. To Total 272 (100.0) 145 127 visualise the efficacy of a molecular marker (i.e., HPV copy number) Age to discriminate two populations (patients who progressed) in the ≤50 140 (51.5) 72 (49.7) 68 (53.5) 0.52 (NS) absence of an arbitrary cut-off value, data were summarised in an >50 132 (48.5) 73 (50.3) 59 (46.5) 16,17 ROC curve. The AUC (area under curve) was calculated as a Histologic subtype fi single measure to discriminate ef cacy. Survival curves were Squamous cell 230 (84.6) 116 (80.0) 114 (89.8) 0.14 (NS) estimated by the Kaplan−Meier method, and compared using the carcinoma log-rank test. For all statistical tests, significance level was defined Adenocarcinoma 27 (9.9) 19 (13.1) 8 (6.3) as p < 0.05. Adenosquamous 10 (3.7) 6 (4.1) 4 (3.1) carcinoma Mixed form or 5 (1.8) 4 (2.8) 1 (0.8) RESULTS undifferenciated Patient characteristics HPV status Clinical, histological, biological (including PIK3CA mutational Genotype 16 155 (57.0) 60 (41.4) 95 (74.8) <0.0001 status) and outcome of the 272 HPV-positive CC patients from Genotype 18 36 (13.2) 28 (19.3) 8 (6.3) the BioRAIDs European study are presented in Table 1. All samples Genotype 45 27 (9.9) 25 (17.2) 2 (1.6) were obtained prior to treatment. Median PFS of the whole cohort Genotype 31 9 (3.3) 6 (4.1) 3 (2.4) was 20.15 months. Fifty-four (20%) patients were treated with Genotype 33 11 (4.0) 4 (2.8) 7 (5.5) upfront surgery, 42 (15%) patients with neoadjuvant chemother- c apy and 176 (65%) patients with external beam radiation therapy Other genotypes 34 (12.5) 22 (15.2) 12 (9.4) with concomitant platinum-based chemotherapy. The majority of Stage FIGO patients (230 patients corresponding to 85%) had squamous cell I/II 205 (75.4) 103 (71.0) 102 (80.3) 0.076 (NS) carcinoma. Classical prognostic biomarkers such as FIGO stage III/IV 67 (24.6) 42 (29.0) 25 (19.7) (2018) and presence of lymph nodes (FIGO III/IV) correlated with Lymph node PFS in the study population (Table 1). Yes 167 (61.4) 101 (69.7) 66 (52.0) 0.0028 = HPV16 was the most common genotype (n 155, 57%) No 105 (38.6) 44 (30.3) 61 (48.0) = = followed by HPV18 (n 36, 13%) and HPV45 (n 27, 10%) Pelvis lymph node (Table 1). Eighty-seven patients (32%) harboured a PIK3CA Yes 165 (60.7) 100 (69.0) 65 (51.2) 0.0027 mutation, which on its own did not correlate to PFS in this subpopulation. No 107 (39.3) 45 (31.0) 62 (48.8) Para-aortic lymph node Integration mechanisms The breakpoints identified on the HPV genome and HPV statuses Yes 43 (15.8) 28 (19.3) 15 (11.8) 0.091 (NS) are reported in Table 2 and Supplementary Table 1. In the absence No 229 (84.2) 117 (80.7) 112 (88.2) of integration (n = 33, 12%), no HPV-chromosomal breakpoint was Initial therapy observed and the viral genome persisted in an episomal form Surgery 54 (19.9) 30 (20.7) 24 (18.9) 0.58 (NS) (EPI). Five HPV integration patterns were observed: 2J-COL (n = 30, Radiotherapy 176 (64.7) 90 (62.1) 86 (67.7) 11%), 2J-NL (n = 53, 20%), 2J (n = 34, 12%), MJ-CL (n = 27, 10%), Neoadjuvant 42 (15.4) 25 (17.2) 17 (13.4) MJ-SC (n = 95, 35%). The BioRAIDs CC series differed significantly chemotherapy from that of HPV-positive anal carcinoma (p < 0.0001) (Fig. 1) PIK3CA mutational d recently published by our team17 in that episomal HPV was much status less frequent in CC as compared to anal carcinoma, while “2J” WT 182 (67.7) 101 (71.1) 81 (63.8) 0.20 (NS) signatures (2J and 2J-NL and 2J-CPL) were more often represented Mutated 87 (32.3) 41 (28.9) 46 (36.2) in CC. The results were similar in HPV16-positive cancers that HPV insertione represent the majority of the subtypes in both cervical and anal EPI 33 (12.1) 9 (6.2) 24 (18.9) <0.0001 cancers (Supplementary Fig. 1). 2J 117 (43.0) 104 (71.7) 13 (10.2) Interestingly, coinfections were observed in 12 CC patients MJ 122 (48.9) 32 (22.1) 90 (70.9) (Supplementary Table 2). These tumours presented unique integration site per HPV genotype, where for each case the HPV Significant results are displayed in bold breakpoints are different. NS not significant. aHPV copy number is a ratio of the number of reads of HPV over control human gene KLK3. Most frequent HPV integration sites b fi Chi-square test. We identi ed >300 different HPV-chromosomal junctions (inter- or cOther genotypes. intra-genic) (Fig. 2 and Supplementary Table 3). The most frequent dInformation available for 269 patients. integration site was in the MACROD2 gene (n = 7) (Supplementary eHPV insertion: EPI episomal, 2J 2 junctions, MJ multiple junctions. Fig. 2) followed by the MIPOL1/TTC6 (n = 5), TP63 (n = 5), and Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 782 several others such as ERBB2 (two sites); KLF12, and RAD51B with a DISCUSSION single site (Fig. 2 and Supplementary Table 3). The two tumours In this CC patient population from the prospective BioRAIDs study, with ERBB2 integration sites were whole-exome sequenced and we were able to identify >300 HPV-chromosomal (inter-genic or both showed ERBB2 amplifications.15 intra-genic) junctions; the MACROD2 gene being the most frequent integration site (n = 7), followed by MIPOL1/TTC6 (n = Association between HPV insertion mechanisms with clinical and 5) and TP63 (n = 5). Interestingly, our data identified a new CC- biological parameters related recurrent integration site in the MACROD2 (mono-ADP- The distribution of HPV integration signatures according to ribosylhydrolase) gene. Non-coding and structural mutations/ clinical, biological and pathological characteristics is presented variations in the germline MACROD2 gene have been associated in Table 2 and Supplementary Table 1. While episomal forms were with psychiatric disorders, obesity and cancer predisposition.18–20 more frequent in PIK3CA mutated tumours (p = 0.023), HPV Deletions in the MACROD2 gene are frequent in colorectal integration signatures were not associated with histological cancer21,22 and are reported to alter DNA repair and sensitivity subtype, with FIGO stage/lymph nodes (presently FIGO stage 3), to DNA damage and consequently impact colorectal tumorigen- or treatment assignation but they were associated with HPV esis.23 Neither RNA expression nor functional studies support a genotype status (p < 0.0001). HPV18 and HPV45 genotypes were tumour suppressor role of MACROD2 gene. This gene spans more always integrated (most frequently as 2J). Multiple (MJ) viral than 2 Mb and constitutes a common fragile site contributing to integration signatures were predominant in HPV16-positive increased genomic instability.24,25 Our results report intronic samples (n = 86/155, 57%) as compared to other HPV genotypes integration sites in the MACROD2 gene yet there is still lack of (n = 36/117; 31%) (Table 2; p < 0.0001). evidence concerning the functional consequence of these intronic integrations within MACROD2. Functional analyses are not Association between the insertion mechanisms and the straightforward due to the high rate of splicing in MACROD2 progression-free survival and the important number of alternative transcripts (coding and There was no significant correlation between the HPV integration non-coding) of variable size. MACROD2 deletions and haploinsuf- signatures (EPI, 2J and MJ) and the PFS (Supplementary Fig. 3a, ficiency were linked to impaired PARP1 activity and chromosomal 3b). Similarly, there was no significant association between the instability in colorectal cancer26 and in liver cancer,27 suggesting a HPV integration signatures and the PFS in the subgroup of HPV16- tumour suppressing function of this gene. Importantly, the positive patients (data not shown). present study identifies HPV integration as a new molecular The most frequent integration site was in the MACROD2 gene pattern of MACROD2 alteration likely causing loss of function, but (n = 7) (Supplementary Fig. 2). Patients with HPV integration sites the seven patients in our cohort with HPV integration in the into the MACROD2 gene ( 5, 6 and 7) did not have a MACROD2 gene are presently insufficient to discern a significantly poorer outcome but the numbers are insufficient to meaningful impact on CC evolution, albeit responsible for draw any conclusions (p = 0.38, Supplementary Fig. 3c). In an genomic instability. exploratory study, interestingly, patients harbouring several viral Previously, frequent integrations in other SCCs were reported in types did not seem to do worse as compared to patients with the MYC, TMEM49, FANCC and RAD51B genes28–30 as well as in the single viral infections (Supplementary Fig. 3d), but this did not following: POU5F1B, FHIT, KLF12, KLF5, HMGA2, LRP1B, LEPREL1, reach statistical significance (p = 0.09). DLG2 and SEMA3D. Slightly less common integration sites were reported in the following genes: AGTR2, DMD, CDH7, DCC, HS3ST4, Comparison of HPV copy number to HPV subtypes, insertion CPNE8, C9orf85, MSX2 and CADM2.9 Several of these previously patterns and outcome reported integration sites into genes such as FHIT, KLF12, RAD51B The HPV copy number was estimated by the ratio of the number were detected in a single or in two patients of the present CC of HPV reads over the control human gene KLK3. The optimal cut- cohort. HPV integration in MIPOL1/TTC6 and TP63 genes were off was four (as determined in the “Methods” section). Patients reported in five patients each. Concordant with our results, were classified into low (ratio < 4, n = 145) vs. high HPV copy Parfenov et al. reported in a head and neck squamous cell number (ratio ≥ 4, n = 127). HPV16-positive patients consistently carcinomas a rearrangement between 3 and 13 had a higher HPV copy number (n = 95/155, 61%) (p < 0.0001) as close to the HPV integration site in a non-coding region but compared to patients with other HPV subtypes (n = 32/117, 27%) involved in a region of chromosome 3 where TP63 genes are (Table 3). Samples with 2J type insertions displayed a low HPV located.31 P63 plays a key role in epidermal keratinocyte copy number while MJ type insertions were associated with a high proliferation and differentiation and is a master regulator of gene HPV copy number (p < 0.0001). Furthermore, patients with a low expression pattern and epigenetic landscape that define epider- HPV copy number showed poor outcome in comparison to mal fate.32 TP63-driven enhancer reprogramming promotes patients with a high HPV copy number (p = 0.011) (Fig. 3). aggressive tumour phenotypes in primary pancreatic ductal

100

75

50

25 High HPV copy number (n=127) p=0.011 Low HPV copy number (n=145)

Progression-free survival (%) survival Progression-free 0 061218 24 Time (months) Fig. 3 Progression-free survival of the 272 HPV-positive cervical cancer patients according to HPV copy number. Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 783 adenocarcinomas.33 HPV integration in TP63 genes was recently AUTHOR CONTRIBUTIONS reported in HPV-positive vulvar cancer patients.34 In another HPV- Conception and design: M.K., M.D., S.M.S. and I.B.; Development of methodology: S.L., positive head and neck squamous cell carcinoma study, HPV sites M.D., E.S.J.; Acquisition of data: A.M., S.V., S.B., G.K., E.S.J., M.P. and C.D.; Analysis and of integration into MIPOL1/TTC6 were identified in more than one interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): tumour sample. The integration of HPV into the ERBB2 gene site M.D., E.G., N.S.; Writing, review, and/or revision of the manuscript: M.K., I.B.; was observed in two patients in association with ERBB2 amplifica- Administrative, technical, or material support (i.e., reporting or organising data, 10 constructing databases): C.L., C.D.; Study supervision: E.M.J.J.B., R.R., W.C., C.L.T., A.N., tions, in concordance with previous reports. I.B. and S.M.S. Twelve percent of CC patients did not display any HPV integration, while 43% had double junctions and 49% multiple junctions’ signatures. The distribution of HPV signatures in our CC ADDITIONAL INFORMATION cohort differed from that previously described in HPV-positive Ethical approval and consent to participate The protocols and results of the anal squamous cell carcinoma with a lower rate of episomal HPV BioRAIDs study [NCT02428842] conducted in 18 European centres have been 17 – as compared to anal cancer (45%). previously published.13 15 A signed informed consent for the participation in the No significant association was observed between HPV integra- study was a prerogative, prior to inclusion and sampling. This study was conducted in tion signatures and treatment type, histological subtype or FIGO accordance with the ethics principles of the Declaration of Helsinki. staging. MJ viral integration signatures were predominant in HPV16-positive samples and tumours with viral integration (2J or Data availability Clinical and whole-exome sequencing data related to the BioRAIDs MJ) had less frequent activating mutations in PIK3CA than those patients were previously published in Scholl et al.15 harbouring episomal HPV, confirming previously reported data.12 Similar results were also observed when considering only HPV16 Competing interests C.L.T. has participated in advisory boards of MSD, BMS, Merck patients (data not shown). This is in accordance with the literature Serono, Astra Zeneca, Roche and Nanobiotix, GSK, Celgene, Rakuten. All other where HPV integration is reported to provide a selective growth authors report no conflict of interest. advantage of cancer cells.6 CC patients with a high HPV copy number had significantly better PFS, as compared to patients with Funding information This project has received funding from the European Union’s low HPV copy number. These results are consistent with other Seventh Program for research, technological development and demonstration under 35,36 grant agreement no. 304810, the Fondation ARC pour la recherche contre le cancer, reports in the literature. ’ In conclusion, while HPV integration is thought to be a random the Association d aide à la recherche en Cancérologie de Saint-Cloud (ARCS), ICGex ANR-10-EQPX-03, and France Génomique ANR-10-INBS-09-08. Funding sources had event, our results point out that some hotspots may impact cancer no involvement in this study and this article. evolution. This would need analyses in larger aggregated datasets. The episomal form of HPV was less frequent in cervical carcinoma Supplementary information is available for this paper at https://doi.org/10.1038/ as compared to another genital carcinoma (anal carcinoma) and s41416-020-01153-4. its presence was significantly associated with high HPV copy number, suggesting a decrease of viral replication upon integra- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims tion. Mutations in PIK3CA were significantly associated with high in published maps and institutional affiliations. HPV copy number and with the episomal form of HPV. The analysis of outcome based on PIK3CA alone did not show an association with poor outcome. In a prior analysis of the BioRAIDS dataset, the association of PIK3CA with epigenetic alterations was REFERENCES associated with a shorter PFS.15 1. Ferlay, J., Soerjomataram, I., Dikshit, R., Eser, S., Mathers, C., Rebelo, M. et al. To our knowledge, this is the first study assessing the Cancer incidence and mortality worldwide: sources, methods and major patterns – prognostic value of HPV integration in a prospectively annotated in GLOBOCAN (2012). Int. J. Cancer 136, E359 E386 (2015). patient cohort and reporting an HPV integration at the MACROD2 2. Schiffman, M. H., Bauer, H. M., Hoover, R. N., Glass, A. G., Cadell, D. M., Rush, B. B. et al. Epidemiologic evidence showing that human papillomavirus infection gene, known to be implicated in impaired PARP1 activity and causes most cervical intraepithelial neoplasia. J. Natl Cancer Inst. 85, 958–964 chromosome instability. (1993). 3. Wentzensen, N., Vinokurova, S. & von Knebel Doeberitz, M. Systematic review of genomic integration sites of human papillomavirus genomes in epithelial dys- ACKNOWLEDGEMENTS plasia and invasive cancer of the female lower genital tract. Cancer Res. 64, We thank Anne de la Rochefordiere, Pierre Fumoleau, Aljosa Mandic, Nina Samet, 3878–3884 (2004). Choumouss Kamoun, Windy Rondoff, Sebastien Armanet, Alexandra Rohel, Souhir 4. Crosbie, E. J., Einstein, M. H., Franceschi, S. & Kitchener, H. C. Human papilloma- Neffati, Marie-Emmanuelle Legrier, Sinette Ngoumou Mabiala, Sylvain Dureau, virus and cervical cancer. Lancet 382, 889–899 (2013). Coralie Errera, Marius Craina, Madalin Margan, Sanne Samuels, Henry Zijlmans, 5. Oyervides-Muñoz, M. A., Pérez-Maya, A. A., Rodríguez-Gutiérrez, H. F., Gómez- Peter Hillemanns, Sorin Dema, Alis Dema, Goran Malenkovic, Branislav Djuran, Macias, G. S., Fajardo-Ramírez, O. R., Treviño, V. et al. Understanding the HPV Anne Floquet, Frédéric Guyon, Pierre Emmanuel Colombo, Michel Fabbro, integration and its progression to cervical cancer. Infect. Genet. Evol. 61, 134–144 Christine Kerr, Charlotte Ngo, Fabrice Lecuru, Eleonor Rivin del Campo, Charles (2018). Coutant, Frédéric Marchal, Nathalie Mesgouez-Nebout, Virginie Fourchotte, Jean 6. Rusan, M., Li, Y. Y. & Hammerman, P. S. Genomic landscape of human Guillaume Feron, Philippe Morice, Eric Deutsch, Pauline Wimberger, Jean-Marc papillomavirus-associated cancers. Clin. Cancer Res. 21, 2009–2019 (2015). Classe, Heiko von der Leyen, Mathieu Minsat, Istvan Nagy, Balazs Balint, Nicolas de 7. Xu, F., Cao, M., Shi, Q., Chen, H., Wang, Y. & Li, X. Integration of the full-length Saint-Jorre, Alexia Savignoni, Franck Perez, Patricia Tresca, Noreen Gleeson, HPV16 genome in cervical cancer and Caski and Siha cell lines and the possible Philippe Hupe, Sergio Roman-Roman, Emmanuel Barillot, Fanny Coffin, Bastiaan ways of HPV integration. Virus Genes 50, 210–220 (2015). Nuijen, Alexandre Boissonnas, Marc Billaud, Laurence Lafanechere, Jaap Verweij, 8. Akagi, K., Li, J., Broutian, T. R., Padilla-Nash, H., Xiao, W., Jiang, B. et al. Genome- Arjan Bandel, Jozien Hellemann, Kirsten Ruigrok-Ritstier, Philipp Harter, Christian wide analysis of HPV integration in human cancers reveals recurrent, focal Kurzeder, Alexander Mustea, Eugeniu Banu, Elisabeta Patcas, Victor Cernat, Andrea genomic instability. Genome Res. 24, 185–199 (2014). Slocker, Michele Mondini, Maud Bossard, Julie Chupin, Sjoerd Rodenhuis, Rene 9. Hu, Z., Zhu, D., Wang, W., Li, W., Jia, W., Zeng, X. et al. Genome-wide profiling of Medema, Anika Havemeier, Thomas Fink, Amelie Michon, Christine Kubiak, Corine HPV integration in cervical cancer identifies clustered genomic hot spots and a Beaufort, Judit Cseklye, Dora Latinovics, Peter Bihari, Anika Havemeier, Isabel Brito, potential microhomology-mediated integration mechanism. Nat. Genet. 47, Bérengère Ouine, Leanne De Koning, Vincent Puard, Elaine Del Nery, Jos Beijnen, 158–163 (2015). Dominique Koensgen, Daniela Bruennert, Slavica Knezevic, Milos Lucic, Alexander 10. Ojesina, A. I., Lichtenstein, L., Freeman, S. S., Pedamallu, C. S., Imaz-Rosshandler, I., Eggermont and Natalja ter Haar for their precious help in the conduct of the RAIDs Pugh, T. J. et al. Landscape of genomic alterations in cervical carcinomas. Nature project. 506, 371–375 (2014). Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 784 11. McBride, A. A. & Warburton, A. The role of integration in oncogenic progression 27. Fujimoto, A., Furuta, M., Totoki, Y., Tsunoda, T., Kato, M., Shiraishi, Y. et al. Whole- of HPV-associated cancers. PLoS Pathog. 13, e1006211 (2017). genome mutational landscape and characterization of noncoding and structural 12. Holmes, A., Lameiras, S., Jeannot, E., Marie, Y., Castera, L., Sastre-Garau, X. et al. mutations in liver cancer. Nat. Genet. 48, 500–509 (2016). Mechanistic signatures of HPV insertions in cervical carcinomas. NPJ Genom. Med. 28. Zhang, R., Shen, C., Zhao, L., Wang, J., McCrae, M., Chen, X. et al. Dysregulation of 1, 16004 (2016). host cellular genes targeted by human papillomavirus (HPV) integration con- 13. Samuels, S., Balint, B., von der Leyen, H., Hupé, P., de Koning, L., Kamoun, C. et al. tributes to HPV-related cervical carcinogenesis. Int. J. Cancer 138, 1163–1174 Precision medicine in cancer: challenges and recommendations from (2016). an EU-funded cervical cancer biobanking study. Br. J. Cancer 115, 1575–1583 29. Zhang, Y., Koneva, L. A., Virani, S., Arthur, A. E., Virani, A., Hall, P. B. et al. Subtypes (2016). of HPV-positive head and neck cancers are associated with HPV characteristics, 14. Ngo, C., Samuels, S., Bagrintseva, K., Slocker, A., Hupé, P., Kenter, G. et al. From copy number alterations, PIK3CA mutation, and pathway signatures. Clin. Cancer prospective biobanking to precision medicine: BIO-RAIDs—an EU study protocol Res. 22, 4735–4745 (2016). in cervical cancer. BMC Cancer 15, 842 (2015). 30. Koneva, L. A., Zhang, Y., Virani, S., Hall, P. B., McHugh, J. B., Chepeha, D. B. et al. 15. Scholl, S., Popovic, M., de la Rochefordiere, A., Girard, E., Dureau, S., Mandic, A. HPV integration in HNSCC correlates with survival outcomes, immune response et al. Clinical and genetic landscape of treatment naive cervical cancer: altera- signatures, and candidate drivers. Mol. Cancer Res. 16,90–102 (2018). tions in PIK3CA and in epigenetic modulators associated with sub-optimal out- 31. Parfenov, M., Pedamallu, C. S., Gehlenborg, N., Freeman, S. S., Danilova, L., Bristow, come. EBioMedicine 43, 253–60 (2019). C. A. et al. Characterization of HPV and host genome interactions in primary head 16. Hanley, J. A. & McNeil, B. J. The meaning and use of the area under a receiver and neck cancers. Proc. Natl Acad. Sci. USA 111, 15544–15549 (2014). operating characteristic (ROC) curve. Radiology 143,29–36 (1982). 32. Soares, E. & Zhou, H. Master regulatory role of p63 in epidermal development and 17. Morel, A., Neuzillet, C., Wack, M., Lameiras, S., Vacher, S., Deloger, M. et al. disease. Cell Mol. Life Sci. 75, 1179–1190 (2018). Mechanistic signatures of human papillomavirus insertions in anal squamous cell 33. Somerville, T. D. D., Xu, Y., Miyabayashi, K., Tiriac, H., Cleary, C. R., Maia-Silva, D. carcinomas. Cancers (Basel) 11, 1846 (2019). et al. TP63-mediated enhancer reprogramming drives the squamous subtype of 18. Lo Re, O., Mazza, T. & Vinciguerra, M. Mono-ADP-ribosylhydrolase MACROD2 is pancreatic ductal adenocarcinoma. Cell Rep. 25, 1741–1755 (2018). dispensable for murine responses to metabolic and genotoxic insults. Front. 34. Thomas, J., Leufflen, L., Chesnais, V., Diry, S., Demange, J., Depardieu, C. et al. Genet. 9, 654 (2018). Identification of specific tumor markers in vulvar carcinoma through extensive 19. Lombardo, B., Esposito, D., Iossa, S., Vitale, A., Verdesca, F., Perrotta, C. et al. human papillomavirus DNA characterization using next generation sequencing Intragenic deletion in MACROD2: a family with complex phenotypes including method. J. Low Genit. Trac. Dis. 24,53–60 (2020). microcephaly, intellectual disability, , renal and pancreatic mal- 35. Deng, T., Feng, Y., Zheng, J., Hg, Q. & Liu, J. Low initial human papillomavirus viral formations. Cytogenet. Genome Res. 158,25–31 (2019). load may, indicate worse proguannosis in patients with cervical carcinoma 20. Hu, N., Kadota, M., Liu, H., Abnet, C. C., Su, H., Wu, H. et al. Genomic landscape of treated with surgery. J. Gynecol. Oncol. 26, 111–117 (2015). somatic alterations in esophageal squamous cell carcinoma and gastric cancer. 36. Lei, J., Ploner, A., Lagheden, C., Eklund, C., Nordqvist Kleppe, S., Andrae, B. et al. Cancer Res. 76, 1714–1723 (2016). High-risk human papillomavirus status and prognosis in invasive cervical cancer: 21. Cancer Genome Atlas Network. Comprehensive molecular characterization of a nationwide cohort study. PLoS Med. 15, e1002666 (2018). human colon and rectal cancer. Nature 487, 330–337 (2012). 22. Andersen, C. L., Lamy, P., Thorsen, K., Kjeldsen, E., Wikman, F., Villesen, P. et al. Frequent genomic loss at chr16p13.2 is associated with poor prognosis in col- Open Access This article is licensed under a Creative Commons orectal cancer. Int. J. Cancer 129, 1848–1858 (2011). Attribution 4.0 International License, which permits use, sharing, 23. Jin, N. & Burkard, M. E. MACROD2, an original cause of CIN? Cancer Discov. 8, adaptation, distribution and reproduction in any medium or format, as long as you give 921–923 (2018). appropriate credit to the original author(s) and the source, provide a link to the Creative 24. Fungtammasan, A., Walsh, E., Chiaromonte, F., Eckert, K. A. & Makova, K. D. A Commons license, and indicate if changes were made. The images or other third party genome-wide analysis of common fragile sites: what features determine chro- material in this article are included in the article’s Creative Commons license, unless mosomal instability in the human genome? Genome Res. 22, 993–1005 (2012). indicated otherwise in a credit line to the material. If material is not included in the 25. Feijs, K. L. H., Cooper, C. D. O. & Žaja, R. The controversial roles of ADP-ribosyl article’s Creative Commons license and your intended use is not permitted by statutory hydrolases MACROD1, MACROD2 and TARG1 in carcinogenesis. Cancers (Basel) regulation or exceeds the permitted use, you will need to obtain permission directly 12, 604 (2020). from the copyright holder. To view a copy of this license, visit http://creativecommons. 26. Sakthianandeswaren, A., Parsons, M. J., Mouradov, D., MacKinnon, R. N., Catimel, org/licenses/by/4.0/. B., Liu, S. et al. MACROD2 haploinsufficiency impairs catalytic activity of PARP1 and promotes chromosome instability and growth of intestinal tumors. Cancer Discov. 8, 988–1005 (2018). © The Author(s) 2020

RAIDS CONSORTIUM Anne de la Rochefordiere16, Pierre Fumoleau17, Aljosa Mandic18, Nina Samet19, Choumouss Kamoun17, Windy Rondoff20, Sebastien Armanet21, Alexandra Rohel21, Souhir Neffati22, Marie-Emmanuelle Legrier17, Sinette Ngoumou Mabiala17, Sylvain Dureau17, Coralie Errera23, Marius Craina24, Madalin Margan24, Sanne Samuels25, Henry Zijlmans25, Peter Hillemanns26, Sorin Dema24, Alis Dema24, Goran Malenkovic18, Branislav Djuran18, Anne Floquet27, Frédéric Guyon27, Pierre Emmanuel Colombo28, Michel Fabbro28, Christine Kerr28, Charlotte Ngo29, Fabrice Lecuru30,31, Eleonor Rivin del Campo32, Charles Coutant33, Frédéric Marchal34, Nathalie Mesgouez-Nebout35, Virginie Fourchotte16, Jean Guillaume Feron16, Philippe Morice20, Eric Deutsch20, Pauline Wimberger36, Jean-Marc Classe37, Heiko von der Leyen38, Mathieu Minsat16, Istvan Nagy39, Balazs Balint39, Nicolas de Saint-Jorre23, Alexia Savignoni17, Franck Perez40, Patricia Tresca41, Noreen Gleeson42, Philippe Hupe16, Sergio Roman Roman17, Emmanuel Barillot17, Fanny Coffin17, Bastiaan Nuijen43, Alexandre Boissonnas44, Marc Billaud45, Laurence Lafanechere45, Jaap Verweij46, Arjan Bandel46, Jozien Hellemann47, Kirsten Ruigrok-Ritstier46, Philipp Harter48, Christian Kurzeder49, Alexander Mustea50, Eugeniu Banu51, Elisabeta Patcas51, Victor Cernat52, Andrea Slocker20, Michele Mondini20, Maud Bossard53, Julie Chupin53, Sjoerd Rodenhuis43, Rene Medema43, Anika Havemeier38, Thomas Fink38, Amelie Michon54, Christine Kubiak54, Corine Beaufort46, Judit Cseklye39, Dora Latinovics39, Peter Bihari39, Isabel Brito17, Bérengère Ouine17, Leanne De Koning17, Vincent Puard17, Elaine Del Nery17, Jos Beijnen43, Dominique Koensgen55, Daniela Bruennert55, Milos Lucic56 and Natalja ter Haar46

16Department of Drug Development and Innovation, Institut Curie, PSL Research University, Paris, France; 17Institut Curie, Paris, France; 18Gynecologic Oncology Department Clinic for Operative Oncology, Institute of Oncology of Vojvodina, Sremska, Serbia; 19Publica Institutul Oncologic, Chișinău, Republic of Moldova; 20Gustave Roussy, Paris, France; 21Arcagy-Gineco, Paris, France; 22Insitut Pasteur, Paris, France; 23Quanticsoft, Paris, France; 24University of Medicine and Pharmacy “Victor Babeș”, Timișoara, Romania; 25Center for Gynaecologic Oncology Amsterdam, Amsterdam UMC and The Netherlands Cancer Institute - Antoni van Leeuwenhoek Hospital, Amsterdam, Netherlands; 26Hannover Medical School, Hanover, Germany; 27Chirurgie onco-gynécologique and Oncology, Institut Bergonié, Centre Régional de Lutte contre le Cancer Bordeaux-Aquitaine, Bordeaux, France; Human papilloma virus (HPV) integration signature in Cervical Cancer:. . . M Kamal et al. 785

28Centre Val d’Aurelle, Paris, France; 29Service de chirurgie cancérologique gynécologique et du sein, Hôpital Européen Georges Pompidou, APHP et faculté de médecine, Université Paris Descartes, Paris, France; 30Department of Surgery, Institut Curie, PSL Research University, 75005 Paris, France; 31Department of Surgery, Institut Curie, PSL Research University, 92210 Saint-Cloud, France; 32Department of Radiation Oncology, Tenon University Hospital, Hôpitaux Universitaires Est Parisien, Sorbonne University Medical Faculty, Paris, France; 33Centre Georges François Leclerc, Paris, France; 34Département de chirurgie, CRAN, UMR 7039, Université de Lorraine, CNRS, Institut de Cancérologie de Lorraine, Vandœuvre-lès-Nancy, France; 35Institut de cancérologie de l’Ouest - site Paul Papin (ICO), Paris, France; 36Department of Gynecology and Obstetrics, Universitätsklinikum Carl Gustav Carus; an der Technischen Universität Dresden, Fetscherstraße 74, 01307 Dresden, Germany; 37René Gauducheau, Paris, France; 38Hannover Clinical Trial Center GmbH, Hannover, Germany; 39SeqOmics Biotechnology Ltd, Vallalkozok utja 7, Morahalom, Hungary; 40Institut Curie, PSL Research University, Paris, France; 41Insitut Curie, Paris, France; 42St James/Trinity College, Dublin, Ireland; 43Netherlands Cancer Institute, Amsterdam, The Netherlands; 44Université Pierre et Marie Curie, Paris, France; 45Université Joseph-Fourier, Grenoble, France; 46Erasmus University Medical Centre, Rotterdam, Netherlands; 47Department of Medical Oncology, Erasmus MC Cancer Institute, Erasmus University Medical Center, 3015 CN Rotterdam, Netherlands; 48Department of Gynecology and Gynecologic Oncology, Ev. Kliniken Essen-Mitte, Essen, Germany; 49Department of Obstetrics and Gynecology, University Hospital of Basel, Basel, Switzerland; 50Department of Gynecology and Gynecological Oncology, University Hospital, 53127 Bonn, Germany; 51Spitalul Sfantul Constantin Brasov, Brasov, Romania; 52Department of Gynecology and Obstetrics, University Medicine Greifswald, Greifswald, Germany; 53Ayming, Gennevilliers, France; 54European Clinical Research Infrastructure Network, Toulouse, France; 55University of Greifswald, Greifswald, Germany; 56Oncology Institute of Vojvodina, Diagnostic Imaging Centre, University of Novi Sad, University School of Medicine, Put Dr. Goldmana 4, Sremska Kamenica, 21204 Novi Sad, Serbia.